Recent Advancements in Autonomous Machine Learning (AML) for Multispectral and Hyperspectral Image Processing of Big Data

Dear Colleagues,

Autonomous machine learning (AML) is an emerging model where the learning system has improved parameters and an improved network architecture. The limitations of the traditional machine learning system has rendered AML to evolve in the technological era. Traditional models rely on static and offline learning. The learning process is begun from the base model with or without the pretrained model structure. The knowledge structure of AML has been constructed automatically through the learning process. For better human–computer interactions, image detection and processing require autonomous learning rather than the recognition of the labeled data. The AML variants handle different types and different magnitudes that drift over time. The stock market, weather forecasting, and online data are applications that drift over time and require the use of AML for effective computation. The application of AML is not limited to regression, clustering, and classification, but also to reinforcement learning. Deep neural networks (DNNs) lack spatial and local contexts. Their self-development is complicated because of the multilayers in DNNs. The categorical entropy is forgotten whenever a layer is introduced in the network architecture. Continual learning and lifelong learning are important research opportunities in regard to AML. The advancements in AML gives us the advantage of adapting to the changing environment and lifelong learning.

HIS and MSI image processing hardly rely on AML for fast and efficient processing. Hyperspectral imaging (HSI) and multispectral imaging (MSI) are applied in the medical field to analyze tissues and medical images for various diagnoses. The multispectral imaging of remote sensing, target detection, and tracking in military applications needs more specific and automated learning. The plant health and ripeness evaluation process is carried out on hyperspectral imaging systems. For the images generated through the high-configuration camera, the robot needs an automated machine learning process to extract the information from the images and preprocess images, such as for image sharpening, restoration, and image segmentation.

The Special Issue aims to bring about the advancement of autonomous machine learning (AML) for hyperspectral imaging (HSI) and multispectral imaging (MSI) in order to better develop AML’s processing and the issues related to the abovementioned problems. We are soliciting original research work on  topics including, but not limited to:

• Novel advancements in AML architecture;
• The role of AML in hyperspectral imaging (HSI) and multispectral imaging (MSI);
• Advanced image segmentation and processing of HSI;
• Hyperparameter tuning for the variance in AML network architecture;
• AML for multispectral imaging and its application in various fields;
• AML for handling the issues of multispectral image processing pattern recognition;
• Continual handling of AML learning problems;
• Multi-stream issues addressed by AML;
• Multispectral imaging and its applications in a new computing model;
• Automated machine learning of new trends and viewpoints;
• Applied image processing for multimedia information security;
• Feature selection/construction and dimensionality reduction in computer vision and image processing;
• Automatic feature extraction and construction in complex images;
• High-performance computing for data processing;
• Artificial neural networks for abnormality detection in images;
• Novel and real-time techniques for compression/decompression of image/video.

Dr. K. Venkatachalam
Dr. Sathishkumar Veerappampalayam Easwaramoorthy
Guest Editors

Manuscript Submission Information

Manuscripts should be submitted online at www.mdpi.com by registering and logging in to this website. Once you are registered, click here to go to the submission form. Manuscripts can be submitted until the deadline. All submissions that pass pre-check are peer-reviewed. Accepted papers will be published continuously in the journal (as soon as accepted) and will be listed together on the special issue website. Research articles, review articles as well as short communications are invited. For planned papers, a title and short abstract (about 100 words) can be sent to the Editorial Office for announcement on this website.

Submitted manuscripts should not have been published previously, nor be under consideration for publication elsewhere (except conference proceedings papers). All manuscripts are thoroughly refereed through a single-blind peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. Mathematics is an international peer-reviewed open access semimonthly journal published by MDPI.

Please visit the Instructions for Authors page before submitting a manuscript. The Article Processing Charge (APC) for publication in this open access journal is 2600 CHF (Swiss Francs). Submitted papers should be well formatted and use good English. Authors may use MDPI's English editing service prior to publication or during author revisions.